( Ixxxiv ) 



the result of an attempt to force upon the theory of natural 

 selection a narrow and rigid interpretation which is not 

 warranted. When we hear that such or such diagnostic 

 characters are too trivial to he explained on the Darwin- 

 Wallace principles, there is underlying such criticism the 

 quite gratuitous assumption that every external character 

 must he in direct and obvious relationship with some 

 external condition of life — that there must be some mani- 

 fest adaptation, such as that between a key and its lock. 

 My object in this address is simply to plead for a relaxation 

 of this rigid interpretation of the term "utility." The 

 theory has always been conceived by me to be not only a theory 

 of obvious adaptations, but also a theory of latent adapta- 

 tions. It is not only a theory of obvious and latent adap- 

 tations, but it is also a theory of obvious and latent 

 adaptability. All those cases of variable protective colour- 

 ing which entomological observation has brought to light, 

 such as the colour adaptability of larvs and pup;B, may 

 be referred to physiological correlation. The adaptability is 

 manifest to us only through its visible token of colour ; of 

 the nature of the physiological processes which have been 

 utilised by natural selection we are as yet ignorant. 

 Poulton's researches have brought us many steps nearer to 

 the intermediate mechanism, but there is yet a long road to 

 traverse before the mode of action of the external stimulus 

 (in this case light) upon the physiological processes can be 

 rendered apparent.* All that is now claimed is an extension 

 of the principle of physiological adaptability from obviously 

 significant characters to those which may be functionless 

 from the point of view of direct utility, but which may, 

 nevertheless, be of supreme importance when regarded as 

 physiological correlates. If such characters are sufficiently 

 constant in the assemblage of individuals constituting a 



* In interpreting results of this kind by the selection theory, it may be 

 aJvisable to point out once again, that "co-adaption" offers no fatal 

 objection. It is not necessary to sujipose that a complex colour-adaptability 

 was developed all at once. There may first liave arisen a responsive power 

 of change towards one colour, and tlien at a later period a similar power for 

 another colour, and so on until, in the most highly perfected species, all 

 possible natural contingencies were provided for. 



